a cli tool i made for myself.

one simple way to transcribe audio from terminal.

all in native python. no subscription. blazingly fast. free. enjoy.

• Supports Many Formats: Works with popular audio (MP3, WAV, M4A) and video (MP4, AVI, MOV) files.
• Automatic Audio Extraction: Extracts audio from video files on the fly using ffmpeg.
• Different Output Options: Save transcriptions in formats like TXT, JSON, SRT, VTT, and TSV.

• Python 3.x
• FFmpeg (for video file support)

pip install aa-whisper

FFmpeg is required if you want to transcribe video files.

Linux:

sudo apt install ffmpeg

macOS:

brew install ffmpeg

Windows (PowerShell):

winget install FFmpeg

Set your Groq API key as an environment variable.

Linux:

echo 'export GROQ_API_KEY="your_groq_api_key_here"' >> ~/.bashrc && source ~/.bashrc

macOS:

echo 'export GROQ_API_KEY="your_groq_api_key_here"' >> ~/.zshrc && source ~/.zshrc

Windows:

[System.Environment]::SetEnvironmentVariable("GROQ_API_KEY", "your_groq_api_key_here", "User")

whisper audio_or_video_file [options]

• --model: Whisper model to use (default: whisper-large-v3-turbo)
• --language: Language code (optional)
• --task: 'transcribe' or 'translate' (default: transcribe)
• --output-dir: Output directory (default: audio_filename_transcription)
• --response-format: 'verbose_json', 'json', 'text', 'srt', 'vtt' (default: verbose_json)
• --version: Show version info

Transcribe audio:

whisper recording.mp3

Transcribe video:

whisper lecture.mp4

Translate to English:

whisper interview.mp3 --task translate

Multiple files at once:

whisper file1.mp3 file2.wav video1.mp4

Each transcription will generate:

• transcript.txt: Plain text
• transcript.json: JSON metadata
• transcript.srt: SubRip subtitles
• transcript.vtt: WebVTT format
• transcript.tsv: Timestamps and text

MIT

• Abdullah Adeeb
• OpenAI Whisper
• Groq API
• FFmpeg